IOSR Journal of Mathematics (IOSR-JM)

e-ISSN: 2278-5728, p-ISSN: 2319-765X. Volume 11, Issue 5 Ver. III (Sep. - Oct. 2015), PP 01-05
www.iosrjournals.org

Cut method and its applications in benzenoid graph

B.Tamilselvi1, K.Ilakkiya2
1

(Associate Professor, Department of Mathematics, PSGR Krishnammal College for Womem, Coimbatore)
2
(Research Scholar, Department of Mathematics, PSGR Krishnammal College for Women, Coimbatore)

Abstract: A general description of the cut method is presented and an overview of its applications in a
benzenoid graph is given. The Wiener index, the Szeged index and the hyper-Wiener index of a benzenoid graph
is calculated based on the consideration of the elementary cuts and the pair of elementary cuts of the
corresponding benzenoid graph.
Keywords: Benzenoid graph, cut method, hyper-Wiener index, Szeged index, Wiener index.

I.

Introduction

Cut method is defined in the following general form. For a given (molecular) graph G,
1. Partition the edge set of G into classes F1 , F2 , , Fk , call them cuts, such that each of the graphs
G Fi , i = 1,2, . . k, consists of two (or more) connected components; and
2. Use properties (of the components) of the graphs G Fi to derive a required property of G.
The cut method can hardly be studied in the above generality; instead we are interested in classes of
(chemical) graphs that allow applicable partitions into cuts and in relevant properties. Often a property of G, that
we are interested in, is some graph invariant, for instance the Wiener index. We could be interested to obtain
expressions for such invariants for certain (chemically) important classes of graphs or to develop fast algorithms
for computing them.
The cut method turned out to be especially useful when if comes to metric properties of graphs. The
key idea how the graphs G Fi can be used to obtain such properties of G is to nd an isometric embedding
f: G H, where H is a properly selected target graph and to use the image f(G) to obtain distance properties of
G. The key subidea is then to select H to be a Cartesian product graph. We introduce and explain the concepts
mentioned in this paragraph in Section 2.
The most prominent class of chemical graphs for which the cut method turned out to be extremely
fruitful is the class of benzenoid graphs. In fact, the 1995 paper [1] and the elaboration of its method for the
computation of the Wiener index of benzenoid graphs (and, more generally, of partial cubes) from [2] can be
considered as the starting point of the cut method. We explain the method and its consequences in Section 3.
A similar approach that works for the Wiener index can be applied to the Szeged index as well. This is
presented in Section 4. We continue with a section on the hyper-Wiener index. Again, the cut method is
applicable; however, in this case it is slightly more involved than the corresponding methods for the Wiener and
the Szeged index because the computation of the hyper-Wiener index requires not only graph distance but also
squares of graphs distances.

II.

Preliminaries

Let G = (V, E) be a connected graph and u, v V. Then the distance dG u, v between u and v is the
number of edges on a shortest u, v-path.
The Cartesian product GH of graphs Gand H is probably the most important graph product and is defined
in the following way:
1. V GH = V(G) V(H);
2. E(GH) Consists of pairs g, h (g , h ) where either g = g and hh E(H), or gg E(G) and h = h .
The graphs Gand H are called factors of GH. The Cartesian product is commutative and associative. The latter
property implies that products of several factors are well-dened. The fundamental metric property of the
Cartesian product is that the distance function is additive:
dGH g, h , g , h = dG g, g + dH (h, h )
This property has been independently discovered several times.
The simplest Cartesian products are products in which all factors are the complete graph on two
vertices K 2 . These graphs are known as hypercubes. More precisely, the n-cube Qn is the Cartesian product of n
factors K 2 , that is,Qn = ni=1 K 2 . It is important to observe that the n-cube Qn can be equivalently described as
the graph whose vertex set consists of all n-tuples b1 b2 bn with bi 0,1 , where two vertices are adjacent
DOI: 10.9790/5728-11530105

www.iosrjournals.org

1 | Page

Cut method and its applications in benzenoid graph

if the corresponding tuples dier in precisely one position.
A subgraph H of a graph G is isometric if for any vertices u and v of H,
dH u, v = dG (u, v)
The class of graphs that consists of all isometric subgraphs of hypercubes turns out to be very
important and has got the name partial cubes. We point out that hypercubes; even cycles, trees, median graphs,
benzenoid graphs, and Cartesian products of partial cubes are all partial cubes.

III.

Wiener index

The Wiener index W(G) of a graph G = (V, E) is defined with

1
W G =
dG (u, v)
2
uU vV
As we have already mentioned, the cut method was rst implemented for a calculation of the Wiener
index of benzenoid graphs. In fact, the method works for any partial cube as the next theorem asserts. For its
formulation we need the following concepts.
Let G be a connected graph. Then e = xy and f = uv are in the Djokovic-Winkler relation [3, 4] if
dG x, u + dG y. v dG x, v + dG (y, u)
The relation is always reflexive and symmetric, and is transitive on partial cubes. Therefore,
partitions the edge set of a partial cube G into equivalence classes, called -classes.
3.1. Theorem
[1] Let G be a partial cube and let F1 , . . Fk be its -classes. Let n1 (Fi ) and n2 (Fi ) be the number of
vertices in the two connected components of G Fi . Then
k

W G =

n1 Fi . n2 (Fi )
i=1

Let G be a partial cube isometrically embedded into Qk . (Note that the number of -classes is equal to
the dimension of the hypercube into G is embedded.) Then a vertex u of G can be considered as a binary k-tuple
u = u1 u2 uk , and the distance between two vertices is the number of positions in which they dier. For
b, b 0,1 , let b, b = 0 if b = b , and b, b = 1 if b b . Having this in mind we can compute as
follows:
1
W G =
dG (u, v)
2
uV vV
1
=
dQ k (u, v)
2
uV vV

1
=
2

(ui , vi )

uV vV i=1
k

=
i=1
k

1
2

(ui , vi )
uV vV

n1 (Fi ) . n2 (Fi )
i=1

In benzenoid graphs the -classes are precisely their orthogonal cuts. Hence if is the set of
orthogonal cuts of a benzenoid graph B, and for we let 1 () and 2 () be the number of vertices in the
two components of , respectively, then Theorem 3.1 specializes as follows [1,2]:
=

1 . 2

We illustrate the use of the above equation in the ovalene , see Fig.1.
The Ovalene 2 has three horizontal cuts, two of them being symmetric. Each of these two cuts
contribute 7.25 to (2 ), while contribution of the remaining cut is 16.16. Hence horizontal cuts contribute
2.7.25+16.16=606. Clearly, there are two more groups of elementary cuts which contributes
2.5.27+2.12.20=750. Hence we conclude that 2 = 606 + 2 . 750 = 2106.
DOI: 10.9790/5728-11530105

www.iosrjournals.org

2 | Page

Cut method and its applications in benzenoid graph

Figure 1: Ovalene and three of its cuts

IV.

Szeged index

For an edge = of a connected graph G let = ()| (, ) < (, ) . The set is

defined analogously. Then the Szeged index of G is defined as:
=

.
()

Now, let be an edge of a partial cube G and suppose that it belongs to the -class. Then it follows
easily from denitions that and induce the connected components of . Therefore, Theorem 3.1
has its variant for the Szeged index:
4.1. Theorem
Let G be a partial cube and let 1 , , be its -classes. Let 1 ( ) and 2 ( ) be the number of
vertices in the two connected components of . Then

. 1 . 2
=1

Theorem 4.1 was elaborated in [5] for benzenoid graphs. Since the -classes of a benzenoid graph are
its cuts, the result specializes to:
4.2. Corollary
[5] Let B be a benzenoid graph and the set of its orthogonal cuts. For let 1 () and 2 () be
the number of vertices in the two components of , respectively. Then
=

. 1 . 2

Consider again the ovalene 2 from Fig.1. The computation of (2 ) goes along the same lines as the
computation of (2 ), except that now we need to multiply each contribution with the size of the
corresponding cut. Therefore,
2 = 4.2.7.25 + (5.16.16) + 2 3. 2. 5. 27 + (4.2.12.20) = 8140.
5. Hyper-Wiener index
The hyper-Wiener index WW was proposed by Randic in [6]. His denition was originally given only
for trees and was extended to all connected graphs = (, ) by Klein, Lukovits and Gutman [7] as follows:
1
1
=
, +
(, )2
(1)
4
4


Note that the rst term is one half of the Wiener index, while in the second we need to compute the
squares of distances. The cut method is applicable also in this case, but because squares of distances are
involved, the method, that we describe next, becomes slightly more involved.
Let G be a partial cube and let 1 , , be its -classes. For each -class let be a representative
of . Then for any 1 < let
11 , = ,
22 , = ,
and
12 , = ,
21 , =
Now consider the following theorem.
DOI: 10.9790/5728-11530105

www.iosrjournals.org

3 | Page

Cut method and its applications in benzenoid graph

5.1. Theorem
[8] Let G be a partial cube with -classes 1 , , and representatives , 1 . Then

= +

11 , . 22 , + 12 , . 21 , Fj

(2)

=1 =+1

The key step in proving Theorem 5.1 is to show that

k

dG (u, v) = 2W G + 4
uV vV

n11 Fi , Fj . n22 Fi , Fj + n12 Fi , Fj . n21 Fi , Fj

i=1 j=i+1

The result then follows immediately by plugging the last equality into (2).
For an example consider again the ovalene.

DOI: 10.9790/5728-11530105

www.iosrjournals.org

4 | Page

Cut method and its applications in benzenoid graph

Figure 2: Types of pairs of cuts in the ovalene

It contains 11 cuts, hence there are

11
= 55 pairs of cuts to be considered. These cuts can be
2

grouped into 17 types that are shown in Fig.2.

There are 2, 1, 4, 4, 4, 4, 4, 4, 4, 7, 4, 2, 4, 1, 2, 2, 2 pairs of cuts in the cases (a), (b), (c), (d), (e), (f),
(g), (h), (i), (j), (k), (l), (m), (n), (o), (p), (q) respectively. Hence the second term of Theorem 5.1 gives
2 7.16 + 0.9 + 7.7 + 0.18 + 4 0.20 + 7.5 + 4 1.14 + 6.11 + 4 3.8 + 4.17 + 4 5.3 + 2.22 +
4 1.12 + 15.4 + 4 4.8 + 12.8 + 4 5.20 + 0.7 + 7 5.12 + 0.15 + 4 5.5 + 0.22 + 2 2.24 + 3.3 +
4 4.19 + 1.8 + 5.12 + 0.14 + 2 12.12 + 0.8 + 2 8.16 + 4.4 + 2 11.11 + 9.1 = 4403.
Since W H2 = 2106 we conclude by Theorem 5.1 that WW H2 = 6509.

V.

Conclusion

The cut method is applied in ovalene (a benzenoid graph) and the Wiener index, the Szeged index and
the hyper-Wiener index has been calculated.

References
[1]
[2]
[3]
[4]
[5]
[6]
[7]

[8]

S. Klavzar, I. Gutman and B. Mohar, Labeling of benzenoid systems which reects the vertexdistance relations, J. Chem. Inf.
Comput. Sci. 35 (1995) 590593.
I. Gutman and S. Klavzar, A method for calculationg Wiener numbers of benzenoid hydro-carbons and phenylenes, ACH Models
Chem. 133 (1996) 389399.
D. Djokovic, Distance preserving subgraphs of hypercubes, J. Combin. Theory Ser. B 14 (1973) 263267.
P. Winkler, Isometric embeddings in products of complete graphs, Discrete Appl. Math. 7 (1984) 221225.
I. Gutman and S. Klavzar, An algorithm for the calculation of the Szeged index of benzenoid hydrocarbons, J. Chem. Inf. Comput.
Sci. 35 (1995) 10111014.
M. Randic, Novel molecular descriptor for structure-property studies, Chem. Phys. Lett. 211 (1993) 478483.
D.J. Klein, I. Lukovits and I. Gutman, On the denition of the hyper-Wiener index for cycle-containing structures, J. Chem. Inf.
Comput. Sci. 35 (1995) 5052.
S. Klavzar, Aplications of isometric embeddings to chemical graphs, DIMACS Ser. Discrete Math. Theoret. Comput. Sci. 51 (2000)
249259.

DOI: 10.9790/5728-11530105

www.iosrjournals.org

5 | Page